FISHERY BULLETIN: VOL. 81, NO. 2 



visibility of fair or better (<4) the total count for the 

 day was estimated by Equation (8a), while for days 

 with visibility of poor or worse (> 4), it was estimated 

 by Equation (8b). (The number of fair or better days 

 recorded each year are listed in Table 3.) 



Verification Experiments 



The distance estimate data from the 1979-80 aerial 

 survey were found to be homogeneous for pooling 

 over sides of the plane, observers, and flight periods 

 (Table 4). Further, the two separate years aerial data 

 were homogeneous, and therefore pooled to form the 

 model distribution (Table 5). This offshore frequen- 

 cy distribution was used as a standard for com- 

 parison with the annual observed distributions, as in 

 Equation (3). 



The values of p ab (Equation (2)), the redistribution 

 parameters calculated from the 1978-79 observer 

 bias experiments, are listed in Table 6. The distance 

 correction factors for each year h{k) indicate that 

 about 20% of passing whales are missed as a function 

 of their distance from shore (Table 7). The cumula- 

 tive proportions of the population estimated to have 

 been observed during the census periods of around 2 

 mo, 8{k), indicate that between 80 and 96% of the 

 population passed the census site during those periods 



TABLE 4.— Analysis of variance for distances of sightings 

 from shore from the 1979-80 aerial transects, with side of 

 plane, period (morning, afternoon), and observer (Reilly, 

 Wolman) as factors. None significant at a = 0.05. 



TABLE 5. — Numbers of whales observed within each of seven dis- 

 tance intervals from shore in aerial transects during 1978-79 and 79- 

 80 with a x 2 test of differences for pooling, x 2 = 5.585 <X 2 6, 0.25. 



Interval 



Distance 

 (mi) 



1978-79 

 no. observed 



1979-80 

 no. observed 



Pooled 

 (%) 



1 



2 

 3 

 4 

 5 

 6 

 7 



Total 



0-0.25 

 0.26-0.5 



06-0.75 

 0.76-1.0 



1.1-1.5 



1.6-2.0 



2.1+ 



14 

 36 

 41 

 74 

 148 

 74 

 29 



416 



1 1 

 33 

 42 

 88 



167 

 61 

 42 



444 



2.91 

 8.02 

 9.65 



18.34 



3663 



15.69 



826 



(Table 7). That is, between 4 and 20% of the popula- 

 tion passed on days before the initiation and after the 

 termination of the annual counting efforts. 



Population Estimates and Variances 



The population size estimate from the 1979-80 cen- 

 sus is 17,577, with a standard error of 2,364. Table 8 

 gives revised population estimates for each of the 13 

 yr using Equation (11), along with raw counts, vari- 

 ances (calculated as outlined in Appendix 2), and 

 95% confidence intervals (C.I.). 



TABLE 6.— Values of the redistribu- 

 tion parameters, p nh , and their 

 variances. 



TABLE 7.— Values of the sighting function h(k), 

 its variance, 6, the cumulative proportion pre- 

 dicted to have passed during the census period, 

 and the variances of 0, for each of 13 annual gray 

 whale censuses. 



TABLE 8.— Raw counts, final population estimates, their variances, 

 standard deviations, and 957c confidence intervals for each of 13 an- 

 nual grey whale censuses. 



Year 



Raw 

 count 



Popu- 

 lation 

 estimate 



Variance 

 (derived) 



SD 



95% confidence 

 limits 



1967-68 

 1968-69 

 1969-70 

 1970-71 

 1971-72 

 1972-73 

 1973-74 

 1974-75 

 1975-76 

 1976-77 

 1977-78 

 1978-79 

 1979-80 



3,077 

 3.265 

 3,399 

 3,264 

 2,667 

 3,684 

 3,889 

 3,836 

 4,295 

 4,720 

 3,717 

 3.927 

 4.924 



13,095 

 11,954 

 12.408 

 11,177 

 10,414 

 14,534 

 14,676 

 13,110 

 15,919 

 16,621 

 14,811 

 13,676 

 17.577 



1,628,883 

 2,388,470 

 2.622,599 

 1,050,782 

 842,864 

 1.817,229 

 2,426,376 

 1,864,729 

 3,545,588 

 3,233.889 

 5,163.965 

 1,270,429 

 5.558,979 



1,276 

 1,545 

 1,619 

 1,625 

 918 

 1,348 

 1,558 

 1,366 

 1,803 

 1,798 

 2,272 

 1.127 

 2.364 



(10,593, 

 ( 8,925, 

 ( 9,234, 

 ( 9,168, 

 ( 8,615, 

 (1 1.892, 

 (11,623, 

 (10,434, 

 (12,228, 

 (13,096, 

 (10,357. 

 (11,467. 

 (11.943, 



15,597) 

 14.983) 

 15,582) 

 13,186) 

 12.213) 

 17,176) 

 17,729) 

 15,786) 

 19,610) 

 20,146) 

 19,265) 

 15.885) 

 22.211) 



276 



